TM
X-MAN and GENESIS
Frequently Asked Questions (FAQs)
What are X-MAN cell models?
X-MAN cell models are controlled pairs (Mutant And Normal) of human, or other
mammalian cell lines, that have defined genetic mutations (X), knocked-in or
knocked-out; and which accurately model the disease causing mutations found
in cancer patients.
What is GENESIS?
GENESIS is a virally-mediated (Adeno-Associated Virus) homologous recombina-
tion technology that enables the first efficient means to perform gene-targeting
in somatic human cell-types. GENESIS has a level of efficiency that is orders of
magnitude greater than previous plasmid-based technologies that are typically
used to create transgenic mice from embryonic stem-cells, but are poorly
efficient in somatic cell types.
How do you make X-MAN cell lines?
GENESIS targets the gene of interest at its endogenous locus; allowing for the first
time the accurate modelling of disease causing mutations (knock-ins or knock-
outs) and single nucleotide polymorphisms (SNPs) in human somatic cell-lines.
GENESIS also permits the definitive study of gene function or protein activity via
highly specific targeted knock-outs of the whole protein or discrete protein
domains, respectively. The steps involved are:
1. Define the mutation to be inserted, or the gene to be knocked-out.
2. Engineer an AAV virus with homology to the target gene, carrying a
defined mutation.
3. Infect the cell line of interest.
4. Clonally select individual cell lines and PCR screen for the correct recombina-
tion event.
5. Confirm the mutant cell line at the genomic level and perform biochemical
characterisation for gain/loss-of-function prior to shipping the cell line.
What are the features of X-MAN cell lines that make them different from other
cell lines?
The most important feature of X-MAN cell-lines is that DNA-modifications are
always made within the endogenous gene; closely recapitulating the genetic
events that lead to a specific disease. The second important feature is that a
matched ‘isogenic’ normal cell-line is also provided; containing a wild-type
version of that gene. This enables the definitive and controlled study of a chosen
genetic alteration on a cell’s function and the search for novel pharmaceutical
agents that selectively target it.
Why do we need to create X-MAN cell lines?
Targeted drug discovery aims to better target the root causes of a disease. In the
case of cancer, there are 100’s of cancer genes now known and very few
cell-based models that either; a) carry your mutant gene or genes of interest, or
b) provide a matched normal control line that enables the study your gene or
genes of interest away from other confounding genetic variations. GENESIS
allows the engineering of any genetic variation into any endogenous target locus
to accurately recreate any target-patient genotype of interest.
How many X-MAN cell lines are there?
To date there are over 150 X-MAN cell lines in our Discovery 1, Discovery 2 and
Researcher panels. These have been assembled by Horizon through in-house
development programs and collaborative research groups using this
gene-targeting technology over the last 15-years. Collectively, these cell lines are
being marketed under the X-MAN brand.
How long does it take to make X-MAN models?
Currently it takes 12 weeks to make a human cell-line pair with a single knock-in
of a desired mutant gene. To make a human cell-line pair with the knock-out of
both target alleles currently takes 16 to 20 weeks.
www.horizondiscovery.com
What is the difference between Discovery 1, Discovery 2 and Researcher X-MAN
lines?
Discovery 1 lines are ‘single-mutant’ gene containing cell-lines - These encom-
pass both ‘gain-of-function’ models (generated via the knock-in of oncogenic
mutant alleles) and ‘loss-of-function’ models (generated via the knock-out of
pre-existing oncogenic mutant alleles or tumour suppressor genes). Discovery 2
lines are ‘double-mutant’ cell-lines - These are currently ‘gain-of-function’ models
generated via the double knock-in of two different oncogenic cancer genes.
X-MAN lines can either be in a normal parental background, or be created from a
cancer cell line with an established mutant/normal genotype. X-MAN Researcher
cell-lines are cancer cell lines (mainly colon) with a specific cancer-associated
gene knocked-out to study its gene-function.
Can X-MAN cell-lines model diseases other than cancer?
Any genetic variation can be introduced using the GENESIS system, so any
genetic disease or predisposition can be modeled using X-MAN cell-lines.
Horizon will be addressing this need for genetically defined disease models in
other therapeutic areas in the coming months and years.
What are AAV-Targeting Vectors?
AAV-targeting vectors are engineered AAV-viruses containing a single-stranded
DNA ‘replacement’ genome that is substantially homologous to the target gene
of interest. DNA-variations incorporated into this AAV- homology construct
enable the efficient introduction of specific mutations or gene-deletions into any
chosen genomic locus; not just within a defined location on chromosome 19,
which represents the natural integration site for wild-type AAV. Engineered
AAV-vectors do not have of any viral genes, and thus, no viral genes are
co-inserted into the target cell’s genome during the homologous recombination
process.
Why are AAV vectors better at gene targeting than other methods?
The homologous recombination machinery is essentially shut off in somatic cells.
However, AAV-vectors uniquely deliver their targeting constructs in the form of a
single-stranded DNA-species, which seems to reactivate this process in some
way – perhaps via mimicking a cross-over event or damaged DNA. While the
mechanism is not well understood, AAV-vectors are consistently more efficient
than double-stranded plasmid- based vectors. Moreover, AAV-vectors elicit
precise alterations within their target genes, without introducing confounding
‘side-modifications’ that are inherent with other gene-targeting techniques (see
‘Zinc Fingers’).
Are AAV viruses pathogenic?
Wild-type Adeno-Associated viruses (AAV) are small non-pathogenic DNA viruses
that infect humans and other vertebrate species without causing a disease.
Wild-type AAV produce a very mild immune response and this property,
combined with their ability to integrate into a specific site on chromosome 19,
has led AAV to be proposed as safe vehicles to perform gene-therapy. Our
engineered AAV-vectors do not have any viral genes, and thus lack the ability to
replicate and no viral genes are incorporated into the final targeted genomic site.
Do any viral ‘sequences’ remain in the cells after causing the desired
mutation(s)?
While there are no viral genes in our AAV-targeting constructs, there are two
non-coding viral ITR-sites that flank the AAV-vector. This allows the vector to be
packaged into a viral capsid when it is introduced into a specialised packaging
cell-line (where all replication and capsid genes provided in trans). Other
elements in the vector are: a) DNA homologous to the target human gene in
question, b) a commonly used heterologous selection marker e.g., NEO, and c)
two lox-P recombination sites to enable the subsequent removal of the selection
marker. In our experience, the non-coding ITR sequences are not integrated into
the final targeted genomic locus due to their position significantly distal to the
site of homologous recombination.
 TM

X-MAN and GENESIS

Frequently Asked Questions (FAQs)

Will GENESIS knockout or modify a gene in my cell line of interest?
Yes. GENESIS has been used to knockout many genes (from tumour suppressor
genes such as p53, PTEN and BRCA) or perform knock-ins of activated mutant
oncogenes (such as K-Ras, PI3K and EGFR) in a range of cell-lines (both human
and mouse). The tropism of AAV is wide with respect to tissue type and species of
cell. The only major requirement is that the cell-line grows continuously in culture
conditions. There is no limit to the number of rounds GENESIS can be used.
Sequential gene-targeting enables either both alleles of a target endogenous
locus to be modified, or the building of multiple disease genotypes within one
target cell.
Should I expect off-target effects with GENESIS?
GENESIS does not use any ‘priming’ DNA-cleavage methodology; homologous
recombination is mediated solely via its approximate 2kb of homology to the
target locus. Potential concerns with random double strand- breaks are therefore
removed. The process of homologous recombination is still competing, however,
with non-homologous recombination; even using our highly efficient
AAV-vectors. For this reason, all our cell-lines go through a clonal selection
process; and Southern Blotting confirms that in each resulting clone, genomic
integrations are always single events using homology-directed AAV-vectors.

How does GENESIS differ from other gene insertion/knockout technologies?

Technique Gene Knock-in Gene Knockout

GENESIS Targeted insertions or modifications are created within
endogenous genes; and so are subject to: a) the correct
gene-regulation mechanisms; and b) accurately reflect
the disease events found in real patients. GENESIS can
introduce subtle point mutations, SNPs as wells as small
insertions with high efficiency. Moreover, many peer
reviewed studies have shown that GENESIS does not
introduce any confounding off-target genomic events.
Gene knockouts are at the endogenous locus, and thus are
definitive, stable and patient relevant. No confounding off-target
effects are elicited at other genomic loci. It requires a 2- step
process:
1. Generate a heterozygous KO
2. Generate a bi-allelic knockout by targeting the second allele.
This process can therefore generate 3 genotypes (+/+; -/+ and -/-);
enabling therefore the analysis of haplo-insufficient gene function.

Plasmid based Insertion is at the endogenous locus and has all the Deletion is at endogenous locus and has all the above benefits, but
homologous above benefits, but it is very inefficient. It also requires a it is inefficient. It also requires a promoterless drug selection
recombination promoterless drug selection strategy entailing bespoke strategy that entails bespoke construct generation
construct generation

Flip-in This is an efficient technique that allows the directed Not applicable
insertion of ‘ectopic’ transgenes at a single pre-defined
genomic locus (via the integration of a FLP recombinase
site). This is not a technique for modifying any endog-
enous locus. Transgenes will usually be under the
control of an exogenous promoter, or a partially defined
promoter-unit in the incorrect genomic location. Their
expression will therefore not be under the same
genomic and epigenetic regulation as the endogenous
loci, which limits the utility of these systems for studying
gene- function. They are however, good for eliciting
rapid and stable exogenous gene expression.

Zinc-Finger ZFNs have been reported to achieve high rates of
Nucleases genetic knock-outs within a target endogenous gene. If
(ZFNs) ZFNs are co-delivered with a transgene construct
homologous to the target gene, genetic knock-in’s or
insertions can also be achieved. However, few peer-
reviewed publications exemplifying this application can
be found and user feedback is mixed regarding the
efficiency of the latter. One worrying aspect is the
potential for ZFNs to produce off-target double-strand
breaks which may also lead to random off-target gene
insertions, deletions and wider genomic instability;
confounding the resulting genotype. Whole genome
sequencing would also be required as a standard
characterization step to determine whether such
off-target genomic alterations are present in any
resulting cell-line.
ZFNs are sequence-directed endonucleases which enable the rapid
and highly efficient (up to 90% in a bulk cell population) disruption
of both alleles of a target gene. However, user- defined or patient-
relevant loss-of-function alterations may be more difficult to
achieve. Off-target deletions or insertions elsewhere in the genome
cannot also be controlled for or readily defined. The speed
advantage of obtaining a biallelic KO in one step is also partially
mitigated if one needs to derive a clonal cell-line study gene
function in a homogenous cell-population. ZFN are likely more
selective than RNAi, however, for high- throughput functional
genomic studies.

www.horizondiscovery.com

X-MAN and GENESIS
Key Publications
Use of isogenic human cancer cells for high-throughput screening and drug
discovery Nat Biotechnol. 2001 Oct; 19(10): 940-5
Targeted transgene insertion into human chromosomes by AAV vectors
Nat Biotechnol. 2002 Jul; 20 (7):735-8
Mutational analysis of the tyrosine kinome in colorectal cancers
Science. 2003 May 9; 300 (5621): 949
Facile methods for generating human somatic cell gene knockouts using AAV
Nucleic Acids Res. 2004 Jan 2; 32(1): e3
Mutational analysis of the tyrosine phosphatome in colorectal cancers
Science. 2004 May 21; 304 (5674): 1164-6
Improved methods for the generation of human gene knockout and knock-in
cell lines Nucleic Acids Res. 2005 Oct 7; 33(18): e158
Somatic mutation of EGFR domain and treatment with gefitinib in colorectal
cancer Ann Oncol. 2005 Nov; 16(11): 1848-9. Epub 2005 Jul 12
Kinase mutations in cancer: chinks in the enemy's armour?
Curr Opin Oncol. 2006 Jan; 18(1): 69-76
Oncogenic activation of the RAS/RAF signalling pathway impairs the response
of metastatic colorectal cancers to anti-epidermal growth factor receptor
antibody therapies Cancer Res. 2007 Mar 15; 67(6): 2643-8
For more information please contact:
Horizon Discovery Ltd.
Building 7300 IQ Cambridge Waterbeach Cambridge CB25 9TL United Kingdom
Tel: +44 (0)1223 655580 - Fax: +44 (0)1223 862240 - Email: info@horizondiscovery.com
© 2010, Horizon Discovery Ltd. GENESIS and X-MAN are registered trade marks of Horizon Discovery Ltd.
www.horizondiscovery.com
TM
Genetic targeting of the kinase activity of the Met receptor in cancer cells
Proc Natl Acad Sci U S A. 2007 Jul 3; 104(27): 11412-7. Epub 2007 Jun 26
Knock-in of Mutant K-ras in Nontumorigenic Human Epithelial Cells as a New
Model for Studying K-ras– Mediated Transformation
Cancer Res. 2007; 67: (18). September 15, 2007
Knock-in of oncogenic Kras does not transform mouse somatic cells but
triggers a transcriptional response that classifies cancers
Cancer Res. 2007 Sep 15; 67(18): 8468-76
Wild-Type BRAF Is Required for Response to Panitumumab or Cetuximab in
Metastatic Colorectal Cancer Journal of Clinical Oncology Published Ahead of Print
on November 10, 2008 as 10.1200/JCO.2008.18.0786
Replacement of Normal with Mutant Alleles in the Genome of Normal Human
Cells Unveils Mutation-Specific Drug Responses Proc Natl Acad Sci U S A Decem-
ber 30, 2008 vol. 105, no. 52
Knock-in of Mutant PI3KCA Activates Multiple Oncogenic Pathways Proc Natl
Acad Sci U S A Printed online at www.pnas.org/cgi/doi/10.1073/pnas/0813351106
PIK3CA Mutations in Colorectal Cancer Are Associated with Clinical Resistance
to EGFR-Targeted Monoclonal Antibodies Cancer Res. 2009; 69: (5). March 1, 2009
A Panel of Isogenic Human Cancer Cells Suggests a Therapeutic Approach for
Cancers with Inactivated p53 Proc Natl Acad Sci U S A Printed online at
www.pnas.org/cgi/doi/10.1073/pnas.0813333106